The government now gets a look at the most powerful AI models before the public does

For twelve days this month, the most capable model OpenAI had built sat behind a government gate. GPT-5.6 was ready on June 26, but the company held it to about 20 vetted organizations at the request of two White House offices, the Office of the National Cyber Director and the Office of Science and Technology Policy, and released it broadly only on July 9. The White House is careful to say it gave no approval, and that release decisions rest with the companies. Still, the most anticipated US model of the year waited on Washington before it reached the public.

The pause traces to Executive Order 14409, signed June 2. It does something narrower than the headlines suggest: the government is not approving AI. The order creates a category, the “covered frontier model,” and asks the labs to hand that model to federal reviewers for up to 30 days before release.

The reviewers, led by the NSA alongside CISA, NIST, and the Defense Department, run the model through a classified benchmark built to test one thing above all: whether it carries advanced cyber or other national-security capabilities that shouldn’t ship unexamined. Clear the benchmark, and the model launches. Raise a flag, and the release slows while the lab and the government work it out. GPT-5.6 was the first real test, and it set the pattern: in the US in 2026, shipping a frontier model now runs through Washington first, whether or not the law says so.

Two words carry most of the weight, and both are unsettled. The first is “covered.” The order doesn’t say, publicly, how capable a model must be to qualify. That threshold is what the classified NSA benchmark is meant to set, and it’s due by August 1, the deadline the order gave agencies to finalize the process and publish the rules. Until then, “covered” means what the reviewers decide, model by model.

The second word is “voluntary.” No company is required to submit anything. But opting out risks federal cloud contracts, government AI procurement, and the light regulatory touch the industry has enjoyed so far. For OpenAI, Google, and Anthropic, the government is not a customer you walk away from. That is how the framework wins compliance without a statute, and why it took hold faster than any AI bill in Congress.

The review is the gentle version. There is a blunt one, too. On June 12, three days after Anthropic released Claude Fable 5, the Commerce Department ordered the company to switch off Fable 5 and its most capable model, Mythos 5, for every foreign user in the world, including Anthropic’s own foreign-national staff, after a jailbroken safeguard was flagged to Washington by Amazon. The models stayed dark for 19 days, until Commerce lifted the order on June 30. That was a mandatory export-control order, not a voluntary review, but it ran through the same office.

Both levers point the same way: past US borders. The framework is also meant to govern who can get frontier models abroad, not only at home; one law-firm analysis calls it a “license to model”. In practice, access to the most capable AI is starting to run through a US review step, wherever the buyer sits.

None of this is settled. The threshold is undefined, the formal rules aren’t published, and “voluntary” could harden or soften with the next fight over mandatory AI regulation. But the direction is clear enough to plan around: for the top tier of AI, the US government has placed itself between the lab and the launch, and it is using procurement, not law, to make that stick.

So what: for the corridor, the catch is foreign access. The same framework that gates US release is also meant to govern who can get these models abroad, which would put Gulf buyers downstream of a US decision. The opening that creates is small but real: tooling that predicts whether a model clears the gate, and non-US model-makers like Canada’s Cohere, positioned as “not subject to a US hold.”

Related

• US grants UAE A:5 status, ending license rules on advanced AI chips — the hardware side of the same corridor valve
• Cohere and Saudi Arabia’s Humain partner on 50MW sovereign AI compute deal — a non-US model-maker as the un-gated alternative